Introducing Replica, a next-generation urban planning tool
De-identified location data can help answer key transportation questions — from a regional level all the way down to a city block.
Who uses the street, in what way, and why? These are common questions that planning agencies consider every day when trying to build better cities. The answers can help them see how well transit is connecting workers to jobs, explore the traffic impact of a new toll lane, or identify the need for bike lanes and wider sidewalks.
But standard planning tools can’t always answer these questions with complete or current details. Too often, planners must rely on costly household surveys conducted years ago or trip counters focused on a single transportation mode. Some agencies have complex modeling software, but that’s often limited by older data and an overly technical interface.
The result is an incomplete sense of city movement patterns and, consequently, a lower confidence in critical transportation and land use decisions.
There’s a key to unlocking better planning tools — right inside the smartphone you might be using to read this article. Our phones have a powerful location awareness that’s transforming many aspects of urban life: helping us get directions, avoid a traffic jam, find a restaurant, or hail a ride. But this type of location data hasn’t widely been used in the service of planning more equitable and adaptable cities.
We believe this powerful data source can help do just that. Meet Replica: a user-friendly modeling tool that uses de-identified mobile location data to give planning agencies a comprehensive portrait of how, when, and why people travel in urban areas.
Replica provides a full set of baseline travel measures that are very difficult to gather and maintain today, including the total number of people on a highway or local street network, what mode they’re using (car, transit, bike, or foot), and their trip purpose (commuting to work, going shopping, heading to school, etc). By updating these measures every three months, Replica also provides the ongoing ability to detect changes in these measures over time — helping planners answer questions about land use and transportation from a regional level all the way down to a city block.
Most importantly, Replica does all that with personal privacy built into its foundation.
A Virtual World With Real Qualities
There are many apps and companies that collect data about your location history and travel patterns via your smartphone. The problem is this data often contains personal information. Replica starts with data that has already been de-identified, meaning we never handle the original, identifiable information. We are not interested in the movement of individuals; we are interested in the collective movement of a particular place.
Replica uses this de-identified data from about 5 percent of the population to learn about travel patterns and create a travel behavior model — basically, a set of rules to represent who’s moving where, when, why, and how. But models aren’t perfect. So we gut check these rules using on-the-ground data (such as manual traffic counts or transit boardings) to make sure Replica is consistent with real-world movement patterns.
We then match these models with what planners often call a “synthetic” population. That’s a very technical term, but the basic idea is that planners can use incomplete samples of census demographic data to create a broad new data set that is statistically representative of the full population. The statistical process also removes any ability to identify a particular individual in the data. (We open-sourced this work last year and encourage others to examine our assumptions or build on top of them.)
When you combine travel behavior models with a representative population, you can confidently replicate trip patterns across a city or metro area.
In Replica, workers go to work and families go out to dinner. Roads are congested at rush hour, downtown sidewalks are busy at lunchtime, and bike paths are full after school. People travel in taxis, on foot, and in carpools. These movements are faithful to real-world activities but not traceable to actual people or specific trips. Planners can use this virtual world to help them make decisions about, and the study the impacts of, transportation or land use — without compromising individual privacy.
From “What Now” to “What If”
Let’s go back to the initial questions — who uses the street, in what way, and why — and consider them through the lens of a city planning agency that wants to make streets safer and friendlier to cyclists. Here’s a look at a Replica dashboard focused on a section of Main Street in Kansas City:
Understanding current conditions. The analysis above shows that nearly 14 percent of all trips in this corridor are made by cyclists and pedestrians, and while most of these people are commuting to work, a notable share are shopping. These baseline counts of trip mode and travel purpose are historically very difficult to gather, but they can help focus planning decisions around empirical evidence. For example, knowing that cyclists and pedestrians are shopping in this area might help demonstrate to local shop-owners that business won’t suffer if street-parking spaces are replaced with a bike lane.
Analyzing changes over time. Currently, there are still few cyclists in this area. But urbanists know that if a model (or, for that matter, a survey) tells you there aren’t many cyclists using a given street, that doesn’t mean people don’t want to bike there — they just might not feel safe enough. The ability to measure changes in usage patterns before and after implementing a bike lane could help planners demonstrate just how many more bike trips a new lane encouraged people to take, making it easier for local officials to support similar interventions elsewhere.
Guiding planning decisions. Over time, we plan to update Replica with the ability to explore prospective service changes and interventions — modeling the impact of Scenario A against Scenario B. We believe this capability can help local officials make the most of limited funding and physical space. It can also help them engage the public around planning decisions in a clearer way. As we’ve written before, transparent models can become the basis for community workshops around things like inclusive street design, helping planners explain the impact that various options might have on different populations.
We are currently building Replica to support the development of plans for Sidewalk Toronto. One of that project’s core objectives is to give communities new tools to adapt much more quickly than cities can today, and we believe Replica can not only help us explore new ideas but to communicate their potential impact to a wider public. As part of this process, we’ll be sharing Replica with local Toronto researchers and public agencies to gather feedback and make it more useful to them.
Later this year, Replica will make its U.S. debut in the Kansas City and Chicago regions, with other areas to follow.
We know models don’t provide simple solutions to planning problems. They’re tools — albeit ones we believe can be more accurate and useful than existing tools. Planning decisions still must reflect the priorities and values of the local community. And many factors beyond modeling outcomes go into urban planning decisions.
But as one Kansas City planner told us during the development of Replica: “The more detail you give me, the more questions I can answer.” By giving planning agencies information that’s more accurate, current, and representative than what’s typically available, we can help them respond more quickly to their community’s needs today — and prepare for the future.
Follow what Sidewalk Labs is thinking, doing, and reading with our weekly newsletter, or on Twitter and Facebook.
